Newtonian Image Understanding: Unfolding the Dynamics of Objects in Static Images
In this paper, we study the challenging problem of predicting the dynamics of objects in static images. Given a query object in an image, our goal is to provide a physical understanding of the object in terms of the forces acting upon it and its long term motion as response to those forces. Direct and explicit estimation of the forces and the motion of objects from a single image is extremely challenging. We define intermediate physical abstractions called Newtonian scenarios and introduce Newtonian Neural Network ($N^3$) that learns to map a single image to a state in a Newtonian scenario. Our experimental evaluations show that our method can reliably predict dynamics of a query object from a single image. In addition, our approach can provide physical reasoning that supports the predicted dynamics in terms of velocity and force vectors. To spur research in this direction we compiled Visual Newtonian Dynamics (VIND) dataset that includes 6806 videos aligned with Newtonian scenarios represented using game engines, and 4516 still images with their ground truth dynamics.
Code (0)
등록된 구현이 없습니다.
Tasks
ObjectSimilar Papers 제목 키워드 기반
Newtonian Scene Understanding: Unfolding the Dynamics of Objects in Static Images
In this paper, we study the challenging problem of predicting the dynamics of objects in static images. Given a query object in an image, our goal is to provide a physical understanding of the object in terms of the forc…
ObjectScene UnderstandingNewtonGen: Physics-Consistent and Controllable Text-to-Video Generation via Neural Newtonian Dynamics
A primary bottleneck in large-scale text-to-video generation today is physical consistency and controllability. Despite recent advances, state-of-the-art models often produce unrealistic motions, such as objects falling …
Text-to-Video GenerationChetaev Instability Framework for Kinetostatic Compliance-Based Protein Unfolding
Understanding the process of protein unfolding plays a crucial role in various applications such as design of folding-based protein engines. Using the well-established kinetostatic compliance (KCM)-based method for model…
Protein FoldingGraph-Based Learning of Free Surface Dynamics in Generalized Newtonian Fluids using Smoothed Particle Hydrodynamics
In this study, we propose a graph neural network (GNN) model for efficiently predicting the flow behavior of non-Newtonian fluids with free surface dynamics. The numerical analysis of non-Newtonian fluids presents signif…
Computational EfficiencyGraph Neural NetworkLagrangians and Newtonian analogues for Biological Systems
This study investigates the potential for biological systems to be governed by a variational principle, suggesting that such systems may evolve to minimize or optimize specific quantities. To explore this idea, we focus …